10 Must-Have Skills for Chinese Content Creators

1|Humanizer-zh A “de-AI-fy” tool for Chinese text. Targets empty clichés, formulaic phrasing, three-part structures, and over-the-top parallelism — the mechanical AI tone — and rewrites it to sound human-written (Chinese localization by Guizang). http://github.com/op7418/Humanizer-zh…

2|dbskill A content “topic diagnostic” tool. Helps you evaluate business copy, polish hooks, break down viral posts, and brainstorm Xiaohongshu titles. http://github.com/dontbesilent2025/dbskill…

3|content-research-writer Covers the entire writing pipeline from topic selection, research, outlining, to first draft. Works for WeChat official account articles, long-form posts, and newsletters. http://github.com/openakita/openakita…(skills/content-research-writer)

4|notebooklm-skill Lets NotebookLM handle the research first, then Claude writes based on your knowledge base — less fabrication, ideal for in-depth articles and industry research. http://github.com/claude-world/notebooklm-skill…

5|khazix-skills An open-source writing collection by Kazik. Heavy on writing workflows, great for AI trend analysis, 10,000-word long-form pieces, and deep opinion pieces. http://github.com/KKKKhazix/khazix-skills…

6|ian-xiaohei-illustrations A body-text illustration tool. Turns ideas, processes, and emotions into “Xiaohei” hand-drawn illustrations — not just a random image thrown at you. http://github.com/helloianneo/ian-xiaohei-illustrations…

7|guizang-social-card-skill Turns finished articles into Xiaohongshu image posts and WeChat cover images. Splits a long article into multiple publishable cards. http://github.com/op7418/guizang-social-card-skill…

8|baoyu-skills A visual toolkit by Baoyu. Produces covers, infographics, structure diagrams, and illustrated explainers — all about packaging your article. http://github.com/jimliu/baoyu-skills…

9|guizang-ppt-skill Turns article ideas into PPTs, presentation visuals, WeChat header images, and Xiaohongshu covers. Makes repurposing content effortless. http://github.com/op7418/guizang-ppt-skill…

10|html-anything Turns Markdown and copy into polished HTML pages, posters, cards, and PNGs. Handles magazine-style layouts and knowledge cards. http://github.com/clockless-org/html-anything…

op7418/Humanizer-zh

Source: https://github.com/op7418/Humanizer-zh

Humanizer-zh: AI Writing De-Tracer Tool (Chinese Version)

Attribution:

• The core files of this project are translated from blader/humanizer (https://github.com/blader/humanizer/tree/main)
• The practical tools section (core rules, quick checklists, quality scoring) draws from hardikpandya/stop-slop (https://github.com/hardikpandya/stop-slop)
• The original project is based on Wikipedia’s Signs of AI writing (https://en.wikipedia.org/wiki/Wikipedia:Signs_of_AI_writing) guide

Project Overview

Humanizer-zh is a tool for removing AI-generated traces from text, helping you rewrite AI-generated content to sound more natural and human-written.

This project is suitable for:

• Editing and reviewing AI-generated content
• Improving the human feel of articles
• Learning to recognize common patterns of AI writing

Installation

Method 1: One-click install via npx (recommended)

npx skills add https://github.com/op7418/Humanizer-zh.git

This is the simplest installation method — it automatically installs the skill to the correct directory.

Method 2: Clone via Git

# Clone to Claude Code's skills directory
git clone https://github.com/op7418/Humanizer-zh.git ~/.claude/skills/humanizer-zh

Method 3: Manual Installation

1. Download the project’s ZIP file or clone it locally

2. Copy the Humanizer-zh folder to Claude Code’s skills directory:

   • macOS/Linux: ~/.claude/skills/
   • Windows: %USERPROFILE%\.claude\skills\

3. Make sure the folder structure looks like this:

   ~/.claude/skills/humanizer-zh/
   ├── SKILL.md       # Skill definition file (Chinese version)
   └── README.md      # Documentation
   

Verifying Installation

After restarting Claude Code or reloading skills, type the following in the conversation:

/humanizer-zh

If installed successfully, the skill will be activated.

Usage

Basic Usage

In Claude Code, you can use Humanizer in the following ways:

1. Directly invoking the skill

/humanizer-zh Please humanize the following text for me:

[paste your AI-generated text]

2. Using it in conversation

Please use humanizer to rewrite this passage to sound more natural:

This project serves as a testament to our team's commitment to innovation. Furthermore, it demonstrates our pivotal role in the ever-evolving technological landscape.

3. Processing file content

/humanizer-zh Please humanize the content in article.md

Example Use Cases

Scenario 1: Rewriting marketing copy

Input:

/humanizer-zh
Nestled in the heart of picturesque downtown Hangzhou, this café boasts a rich cultural heritage and breathtaking decor. It serves as a focal point of the city's coffee culture, providing customers with a seamless, intuitive, and vibrant experience.

Sample Output:

This café has been open in downtown Hangzhou for three years, known for its pour-over coffee and a space renovated from an old building.

Scenario 2: Rewriting an academic abstract

Input:

/humanizer-zh
This study delves into the critical role of machine learning in medical diagnostics, highlighting its importance in the ever-evolving healthcare landscape. Furthermore, it lays a solid foundation for future developments in the field.

Sample Output:

This study analyzed the application of machine learning in medical diagnostics, focusing on early lung cancer screening. The research used medical record data from 5,000 cases between 2019 and 2023.

Scenario 3: Rewriting a blog post

Input:

/humanizer-zh
Artificial intelligence is not just a technology; it is a revolution in the way we think about the future. Industry experts believe it will have a lasting impact on society as a whole.

Sample Output:

I’ve been thinking about how AI will change the way we work. Last week, I chatted with a few friends who build products — some are excited, some worry about job loss. The truth is probably somewhere boring in between.

AI Writing Patterns Detected

This tool can identify and fix 24 types of AI writing traces, grouped into four categories:

📝 Content Patterns (6)

1. Overemphasizing significance, legacy, and broader trends
2. Overemphasizing prominence and media coverage
3. Shallow analysis ending with “-ing” constructions
4. Promotional and advertising-style language
5. Vague attribution and hedging language
6. Outline-style “Challenges and Future Outlook” sections

🔤 Language and Grammar Patterns (6)

7. Overused “AI vocabulary”
8. Avoiding the verb “to be” (copula avoidance)
9. Negative parallelism
10. Overuse of the rule of three
11. Deliberate word substitution (synonym rotation)
12. False scope

🎨 Style Patterns (6)

13. Overuse of dashes
14. Overuse of bold text
15. Vertical lists with inline headers
16. Title case in headings
17. Emojis
18. Curly/smart quotes

💬 Communication Patterns and Filler Words (6)

19. Traces of collaborative conversation
20. Knowledge cutoff disclaimers
21. Flattering/servile tone
22. Filler phrases
23. Excessive qualification
24. Generic positive conclusions

File Description

• SKILL.md - Chinese version of the skill definition file
• README.md - This documentation

Note: For the original English version, refer to blader/humanizer (https://github.com/blader/humanizer)

Manual Usage Guide

Basic Workflow

1. Identify AI patterns - Scan the text against the 24 patterns listed in SKILL.md
2. Rewrite problematic sections - Replace AI traces with natural expressions
3. Preserve core meaning - Ensure information integrity
4. Maintain appropriate tone - Match the style the text should have
5. Inject genuine personality - Give the writing a “human touch”

Key Principles

✨ Aim not just for “clean,” but for “alive”

Avoiding AI patterns is just the baseline; good writing needs a real human voice:

• Have opinions - Don’t just report facts; react to them
• Vary the rhythm - Mix long and short sentences
• Acknowledge complexity - Real people have complicated feelings
• Use “I” when appropriate - First-person is honest
• Allow some messiness - Perfect structure reads as mechanical
• Be specific about feelings - Replace abstract generalizations with concrete details

Example Comparison

Before (AI-flavored):

The new software update serves as a testament to the company’s commitment to innovation. Furthermore, it provides a seamless, intuitive, and powerful user experience — ensuring users can accomplish their goals efficiently. This is not merely an update, but a revolution in the way we think about productivity.

After (Humanized):

The software update adds batch processing, keyboard shortcuts, and offline mode. Early feedback from beta testers has been positive, with most reporting faster task completion.

Changes:

• Removed exaggerated symbolism (“serves as a testament to”)
• Removed AI vocabulary (“furthermore,” “seamless”)
• Removed the rule of three (“seamless, intuitive, and powerful”)
• Removed negative parallelism (“not merely… but…”)
• Added specific features and real feedback

Common AI Vocabulary Warning List

The following words appear with abnormally high frequency in AI-generated text:

• Furthermore, crucial, delve into, emphasize
• Enduring, enhance, foster, gain
• Highlight, interplay, complex/complexity
• Landscape (abstract noun), pivotal, showcase
• Tapestry (abstract noun), testament, underscore
• Valuable, vibrant

Contributing

If you find translation issues or want to improve the documentation, feel free to submit an Issue or Pull Request.

Chinese Context Specificities

During translation and adaptation, we considered the characteristics of Chinese writing:

• Some English patterns manifest differently in Chinese (e.g., title capitalization)
• Added examples suited to the Chinese context
• Adjusted certain expressions to align with Chinese conventions

Reference Resources

• Wikipedia: Signs of AI writing (https://en.wikipedia.org/wiki/Wikipedia:Signs_of_AI_writing) - Original source guide
• WikiProject AI Cleanup (https://en.wikipedia.org/wiki/Wikipedia:WikiProject_AI_Cleanup) - Wikipedia’s AI cleanup project
• blader/humanizer (https://github.com/blader/humanizer) - Original English project
• hardikpandya/stop-slop (https://github.com/hardikpandya/stop-slop) - Inspiration for the practical tools section

License

This translation project follows the license of the original project. Core content is based on observations and summaries from the Wikipedia community.

Tip: This tool is not meant to “deceive” AI detectors, but to genuinely improve writing quality. The best way to “de-AI” text is to give it real human thought and voice.